Manager Biomanufacturing Sciences Network - DCVMN

QbD for Tangential Flow Filtration

Tathagata RayManagerBiomanufacturing Sciences Network

Agenda

QbD Concept1

Review of TFF 2

Key Application of TFF in common vaccines3

4 QbD workflow

Quality is a key regulatory concern

Efficacy/Strength

Identity & Purity

Safety

Does the production process result in product/residues that interfere with final

product strength or efficacy?

Does the production process result in product/residues that interfere with final product

purity?

Does the production process result in product/residues that are toxic to the

patient?

Suitability of either a drug substance or product for its intended use (ICH Q6A)

Quality in Biopharmaceuticals: Where we stand …….. Where we intend to …….

1

100

10000

1000000

2σ 4σ 6σ

DPMO

Restaurant bills

Airline baggage arrival

Aviation industry

Nuclear industry

How do we bridge this quality gap ? ……. By testing !!!

Or, it has to be built in by design ….

DPMO – Defects per million opportunities

QbD …. Overall approach

CQA 4

CQA 2CQA 1

CQA 3

PRODUCT DESIGN SPACECLINICAL DESIGN SPACE PROCESS DESIGN SPACE

TARGET PRODUCT PROFILE

CONCEPT > DESIGN > PRE-CLINICAL > CLINICAL > MASS PRODUCTION

Animal studies

PROCESS PARAMETERS

QbD: From process ….. To step ……

Quality target product profile

Identify CQAsDesign Process

Design formulation

Material attributes and Process parameters

CQAint1 CQAint2

Identify and control sources of variability

Monitor and update

Process Step Characterization

based on risk assessment / control

ie – TFF

Basic UF Applications & Schematic

ClarificationClarification

Product passes through the membrane

Larger particles / molecules retained by membrane

ConcentrationConcentration

Product retained by the membrane Solvent (buffer) passes through the

membrane

Diafiltration (Buffer Exchange or contaminant removal)

Diafiltration (Buffer Exchange or contaminant removal)

Product retained by the membrane Solvent (buffer) passes through the

membrane, new solvent added to product

Contaminant removal

Pressures and Flows in UF Membranes

8

PF = feed pressure [bar or psi]PR = retentate pressure [bar or psi]Pf = filtrate pressure [bar or psi]= osmotic pressure [bar or psi]

QF = feed flow rate [L h-1]

QR = retentate flow rate [L h-1]

Qf = filtrate flow rate [L h-1]

k = mass transfer coefficient [L/m2*h]

Cb = protein concentration in bulk solution [g L-1]

Cw = protein concentration at membrane [g L-1]

Cf = protein concentration in filtrate [g L-1]

Cb

Cw

Cf

MembraneMembrane

MembraneMembrane

QF

PF

QR

PR

Qf

Pf

Typical TFF UF applications

• Single pump / Retentate control

• Low permeability / High TMP

• TMP / delta P controlled

• Optimization of Feed flow / TMP / Diafiltration strategy

D ia filtr a t io n B u ffe r

F e e d T a n k

Q P

Q R

F e e dP u m p

F iltr a t io nM o d u le

P F

P R

R e te n ta te V a lv e

Q F

P P

F E E D

R E T E N T A T E

P E R M E A T E

0

20

40

60

80

100

120

140

160

180

200

0 10 20 30 40 50

TMP [psid]

Flux

[LM

H]

Feed Flux = 5 L/min/m2

Feed DP = 20 psi

Feed Flux = 3 L/min/m2

Feed DP = 10 psi

Optimum Point:TMP = 30 psidJf = 150 LMH

Optimum Point:TMP = 25 psidJf = 86 LMH

Adenovirus vaccine: Typical UF TFF process parameters

Purification: last UF/DF Step

■ Success Criteria– Good Yield & Retention

– Contaminant removal ► RNA

■ Solution– No permeate control

– Pellicon® 2 Biomax ® or Ultracel ® 100 or 300kD, C screen

Parameters Biomax/Ultracel 100-300 KD

Feed flow (l/min/m²) 4-8

TMP (bar) 0.3-1

Average flux (LMH) 25-50

Volumetric Concentration Factor

4-10

Diafiltration volume 5-12

Diafiltration Buffer

Feed Tank

Q P

Q R

FeedPump

FiltrationM odule

P F

P R

Retentate Valve

Q F

P P

FEED

RETEN TATE

PERM EATE

Typical TFF MF / Open UF applications

Set up of an equipment with permeate control

■ 2 Pump System / Flux controlled

■ Permeate Valve & Flow Meter

P-feed

Retentatevalve

P-perm

P-ret

Recirculation or Feed pump

Permeatepump

Pump forproduct addition (Fed batch)

or Diafiltration buffer

Recovery

Processtank

Permeatetank

Temp°C

Productor buffer

tank

Viral Antigen: Egg-based Influenza Vaccine Typical MF / Open UF TFF process parameters

Parameters Biomax 1000kD

Feed flow (l/min/m²) 6

TMP (bar) <0.3

Initial flux (LMH) 30

Final flux (LMH) 30

Average flux (LMH) 30

Volumetric Concentration Factor 10

Diafiltration volume 2

Purification:Success Criteria

– Good yield & Retention

– Higher purity and Contaminant Removal► Ovalbumine

Solution

– Permeate control

– Pellicon ® 2 Biomax ® 1000, V Screen

Result

– Retention > 99.99%

– Contaminant removal > 75%

P-feed

Retentatevalve

P-perm

P-ret

Recirculation or Feed pump

Permeatepump

Pump forproduct addition (Fed batch)

or Diafiltration buffer

Recovery

Processtank

Permeatetank

Temp°C

Productor buffer

tank

With or Without Permeate Control

Contaminant removal and Antigen retention Vs. VCF under different set of operating conditions

0

10

20

30

40

50

60

70

80

1 2 3 4 5 6 7 8Volumetric Concentration Factor

Con

tam

inan

t re

mov

al (%

)

Contaminant removalOptimum operating conditions

98

98.2

98.4

98.6

98.8

99

99.2

99.4

99.6

99.8

100

Antigen retention (%

)

Antigen retention

0

2

4

6

8

10

12

14

16

Contaminant removalNon-optimum operating conditions

Optimum operating conditions: feed flow = 6 lpm/m², TMP< 0.4 bar, permeate controlled at 30 LMHNon-optimum operating conditions: feed flow = 6 lpm/m², TMP> 1 bar, no permeate control

Case study results: Clearance of Benzonase® digested DNA across diafiltration with Pellicon® 2 Biomax® 300 kDa

1 2 3 4 5 6 7 8 9 Various UF samples

■ Lane 1 – Marker (100 BP)

■ Lane 2 – Undigested DNA in Feed

■ Lane 3 – After Benzonase® digestion

■ Lane 4 – Post Recirc retentate

■ Lanes 5, 6, 7, 8 – Retentate samples after 1, 3, 5, 8 DV

■ Lane 9 – Permeate at 5DV

Benzonase ® can also be effectively removed with diafiltration

Or

Can also be removed in subsequent chrom operations

TFF and QbD

TFFWell known mass transfer fundamentals

Proven engineering principles and design equations

Mechanistic understanding of hydrodynamic principles governing separation process

Identifies “Optimal operating conditions” and define “Design space”QbD

Principles of boundary layer flow, Fickian

diffusion, Darcy pore flow, Film theory

Eddy diffusivity model, turbulent/laminar flow, Reynolds/Schmidt/Sh

erwood numbers

Navigator 1Define and agree on the Quality Target Product Profile (QTPP)

Determine the Critical Quality Attributes (CQAs)

Link raw material and process parameters to the CQAs

Risk assessment to identify Critical Process Parameters (CPP)

Define the Design Space (DOE study)

Define the Control Strategy to keep process within design space.

Product lifecycle management and continual improvement

TFF step attributes - VaccinesFunction Performance Specs Attributes

Product (CQA)

Performance(KPA)

Harvest Concentration (dewatering), Purification LRV (HCP, NA), Clarification LRV (Turbidity), Yield, Cost, Time

Purification LRV (HCP, NA)

Concentration(dewatering), Clarification LRV (Turbidity), Yield, COGs, Time

Purification / Fractionation

Purification LRV (HCP, NA, Benzonase®, Conjugation reagents, ADH), Yield, Cost, Time

Purification LRV (HCP, NA, Benzonase®, Conjugation reagents, ADH)

Yield, COGs, Time

Formulation UFDF final concentration, buffer composition, LRV process extractable, Yield, Cost, Time

UFDF final concentration, buffer composition, LRV process extractable

Yield, COGs, Time

Examples of UF product CQAs in a Vaccine process

Vaccine Product CQA Process templateInfluenza Contaminant removal

(Ovalbumin, Ovotransferrin, Ovoglobulins, Lysozyme and others)

Product conc./ quality

KPA – Yield, COGs, Time

VectoredVaccine (Malaria, Dengue)

Benzonase® / NA removal

Product conc./ quality

Buffer change


Purification

Formulation

Purification Formulation


Vaccine Product CQA Process template

Polysaccharide Conjugate Vaccines

(Pneumonia,Meningitis, Influenza)

Product conc. Purity

Buffer exchange

ADH removal from PRP - ADH mass

Fractionation of PRP-ADH-TT complex from unreacted mass

Removal of reaction chemicals


Concentration

PurificationPRP-ADH

Diafiltration

FractionationPRP-ADH-TT


Vaccine Product CQA Process template

Live Attenuated Viral Vaccines

(MMR, Dengue, JE, Oral polio)

Product conc. Purity

Buffer exchange

Benzonase® / NA removal

KPA – Yield, COGs, Time, Operator safety (Clarification)

Clarification

Purification Formulation

Link Material attributes and Process parameters to CQA & KPA

Process(TFF)

RawMaterial*

ProcessParameters

Retentate(Attributes)

• Product concentration, • Product purity, • Osmolality, pH • Contaminant removal• Fractionation profile

• Yield/retention • COG’s, (Membrane reuse)• Time*Raw Material attributes: Feed solution, buffers, filters

Overview of process and material attributes

22

VrCr

VpCp

Cf

MembraneMembrane

MembraneMembrane

QF QR

Qf

TMP

Rg

Rm

Mass Transfer coefficient (k) = 1 / Rg+Rm

R = 1- Cp/Cr

Overall R(retention) depends on membrane retention (R90) / gel layer

QF, Feed, TMP

Membrane

Rproduct High / Rcontaminant Low – when Product in Retentate

VfCf

X (adsorptive loss) = (CfVf – CrVr – CpVp) / A

Y (% product loss) = 100 X [ 1- { e (Vp/Vf)X(1-R) }-1]

Jfiltrate = k X TMP = [1/Rg+Rm] X TMP

Link Material attributes and Process parameters to CQA & KPACQA

• Product concentration, • Product quality, • Osmolality, pH • Contaminant removal• Fractionation profile

Membrane characteristics (MOC / R value)

Device characteristics (screen / Channel)

Feed characteristics

Shear – Quality

Diafiltration (contaminant removal / purity / exchange)

TMP

Feed flow / Cross flow

Cleaning

KPA

• Yield/retention • COG’s, (Membrane reuse)• Time

Risk assessment

Scoring of Process Parameters and Quality (& Process) Attributes

Cumulative score = ∑ (Impact of parameter x Weight of quality attribute)

TFF Risk Analysis – Example of a UF-DF in typical formulation application

Attribute Weight 10 5 5 10 7 10

Feed Flow Rate (LPM/m2) 1 5 5 7 5 1 175Transmembrane Pressure (psi) 1 5 10 5 5 1 180Process Loading, L/m2 5 5 7 5 1 1 177No of DiaVolumes 1 1 7 5 1 10 207Feed characteristics (Titre) 1 1 1 5 5 1 115Recovery, L/m2 10 1 1 1 1 1 137Membrane characteristics 5 1 7 1 1 1 117Temperature 1 1 7 5 1 1 117

Score

Step

Yield

Mem

brane Re

use

Process Time

Product a

ggregatio

n (quality)

Product titre

(Retentate conc)

Conc/Diaf

Transfer to

form

ulation

bufferPhase Parameter

Process Attribute Product Attribute

TFF Risk Analysis – Example of a UF-DF in typical purification application



Score

Step

Yield

Mem

brane Re

use

Process Time

Product q

uality

Product titre

(Retentate conc)

Conc/Diaf

Removal of

contam

inants (e

g ‐

Albumin / Benzonase /

Phase Parameter


TFF Risk Analysis – Example of a UF-DF in typical fractionation application



Conc/Diaf

% product

Phase Parameter


Step

Yield

Mem

brane Re

use

Process Time

Product a

ggregatio

n (quality)

Product titre

(Retentate conc)

Score

TFF Risk Analysis – Example of a UF-DF in a Influenza formulation - Sample experimental design for DOE characterization

Parameter Low Middle High Score Scientific Rationale

Feed Flow Rate (LPM/m2) 3 5 7 175 Impacts polarization, fouling & potentially quality

Transmembrane Pressure (psi) 2 4 6 180 Impacts flux (time), polarization and fouling

No of Diavolumes 6 10 14 207 Impacts time and buffer exchange efficiency

Loading, L/m2 20 30 40 177 Potential impact on process time, COGs

Feed Stock 2-3 lots representing variability in Feed

Linking Variable

Potential for variability in feed titre, or impurity levels to impact UF-DF performance

Filter2-3 lots representing variability in Membrane characteristics

Linking Variable

Variability in membrane retention can potentially influence yield, contaminant removal

QbD Concept - Process Parameters

Process parameters, whose variability has an impact on a CQA, need to be monitored and controlled

30

Critical Process Parameter (CPP)1

Critical Process Parameter (CPP) Variability in CPP has an impact on critical quality attribute and therefore should be

monitored or controlled to ensure process produces the desired quality. A CPP has a high risk of falling outside the design space.

2 Well Controlled Critical Process Parameter (WC-CPP) Variability in CPP has an impact on critical quality attribute and therefore should be

monitored or controlled to ensure process produces the desired quality. A CPP has a low risk of falling outside the design space.

3Key Process Parameter (KPP) An adjustable parameter (variable) of the process that, when maintained within a narrow

range, ensures operational reliability. A key process parameter does not affect critical quality attributes.

4General Process Parameter All “other” parameters

A Hypothetical Summary of Process Parameter Classification and Ranges - Formulation

31

In a “Purification” application “Diafiltration” can be “CPP” based on risk assessment wrt CQA

QbD Concept - Design Space

33

Design Space

Defined as: “the multidimensional combination and interaction of input variables (e.g., material attributes) and process parameters that have been demonstrated to provide assurance of quality.” ICH Q8(R2), http://www.fda.gov/downloads/Drugs/Guidances/ucm073507.pdf

Demonstrated range of all process parameters where process meets the CQAs Consists of Knowledge space, design space and control space

“Implementation of Quality by Design”. J.F. Haury, Amgen 2006 http://www.fda.gov/downloads/AboutFDA/CentersOffices/CDER/ucm118776.pdf

CQA and Process design space

INPU

T (X

)

OU

TPU

T (Y

)Raw material

People

Process parameters

Y = f (X)

CQA

PP1

PP2

CQA1

CQA 2

Relationships:CQA1 = function (RM1, RM2, PP1)

CQA2 = function (PP1)

PP2 might not be needed in the establishment of design space

VARIABLE INPUT ADAPTED PROCESS CONSISTENT OUTPUT

RM1

RM2

Example of Experiments to Determine Acceptable Range of Process parameters

0

5

10

15

20

25

30

0 0.5 1 1.5

Flux

(LMH)

TMP (bar)

2L/min/m2

4L/min/m2

6L/min/m2

8 L/min/m2

Typical flux excursion experiments have two variables and an output.• Variables – feed flow rate/cross flow rate and

TMP• Output - Flux

Parameter Ranges

Challenges in setting Process Parameter Ranges in DOE studies

36

If too wide Everything becomes critical

If too narrow Nothing is critical

“Right “ Range Meaningful determination of critical

Prior knowledge & platform parameter information serve as good guidance

Alert and Action Limits

37

Target

Target Range

Alert Limit

Alert Limit

Acceptable Range(License Range)

Observation Observation

Action Limit

Discrepancy

Action Limit

Discrepancy

Parameter range - UF Example

38

Design Space and Control Strategy

40

Design Space

Process(TFF)

RawMaterial

ProcessParameters

Retentate(Attributes)

Control

Monitoring Process Parameters

Parameter Monitoring & Control: UF / MF Examples

41

Feed

Filter

Feed Flow, TMP

Concentration End-Point

Diafiltration End-PointMaintain diafiltration flow rate = permeate flow rate through

retentate level control. Time or permeate volume measurement. High end analytical support in case of purification /fractionation

Feed concentrations of product, impurities and buffer components can be measured directly and/or controlled through

the previous step.

Filter properties such as retention, permeability – monitor through vendor quality audit.

Control the feed pump flow using a mass flow meter & PID control. Use retentate / permeate flow control valve & pressure transmitters (feed, retentate, permeate) to control TMP (PID).

Retentate tank volume (Wt) or level specification.

Control of Process Parameters

■ Ensure product quality and safety (for CPPs)– Control within design space to ensure consistent product quality

and process performance

■ Ensure that the commercial manufacturing process is consistent and robust (KPPs)– Also, controlled within target range to ensure consistent process

performance ► Non CPPs need to be controlled just as much as CPPs do

42

Control of Process Parameters■ Control Strategy

– A control plan derived from current product and process understanding that assures product quality and process performance

– A method to keep or maintain the ‘process’ within the design space.

43

Summary

■ QbD – represents a scientific approach to build-in & ensure quality in drug products– Emphasizes process understanding, relationship between CPPs, CQAs,

QTPPs using a methodical approach (risk assessment)

■ QbD principles may be applied to TFF to determine the important process parameters

- Feed flow, TMP (flux), Diavolumes can be CPPs or KPPs

■ Process control strategies help ensure that process parameters are maintained within the desired range to ensure product quality and reliable process operation

44

Tathagata [email protected]

Manager Biomanufacturing Sciences Network - DCVMN

Documents

Transcript of Manager Biomanufacturing Sciences Network - DCVMN